Apnea-hypopnea events during REM/non-REM sleep and hypertension among patients with obstructive sleep apnea

R. Ren,Y. Zhang,L. Yang,L. Tan,T. Li,X. Tang
DOI: https://doi.org/10.1016/j.sleep.2019.11.893
IF: 6.313
2019-01-01
SLEEP
Abstract:Obstructive sleep apnea (OSA) has been associated with hypertension. It is possible that the association between OSA and hypertension is distinct for non-REM versus REM sleep because of differences in sleep-state-dependent sympathetic activation and/or degree of hypoxemia. We intended to examine the association between REM-related/non-REM-related OSA and hypertension, and whether the severity of OSA modifies such association. A total of 10,102 patients with apnea-hypopnea index (AHI) ≥ 5/h were recruited into this study (83.8% males, mean age = 44.92 ± 11.86 years). Hypertension was defined based either on direct blood pressure measures or on physician diagnosis. OSA severity during REM and non-REM sleep was quantified using the apnea-hypopneaindex in REM (AHIREM) and non-REM sleep (AHINREM), respectively. Linear regression was used to assess the associations of AHIREM and AHINREM with hypertension and blood pressure. A 53.4% was found to have hypertension in total observed OSA patients. After adjusting for age, gender, body mass index (BMI), Epworth sleepiness scale, tobacco use, alcohol use, nocturnal oxygen desaturation, sleep duration and efficiency, AHIREM was only associated with diastolic blood pressure (DBP), while AHINREM was associated with percentage of hypertension, systolic blood pressure (SBP) and DBP. Among mild-moderate OSA patients, AHIREM was associated with DBP (B = 0.87; 95% CI, 0.05-1.58; P = 0.037) while AHINREM was not; among severe OSA patients, AHINREM was associated with percentage of hypertension (B = 0.28; 95% CI, 0.18-0.37; P < 0.001), SBP (B = 7.49; 95% CI, 4.53-10.44; P < 0.001) and DBP (B = 7.23; 95% CI, 5.07-9.38; P < 0.001) , whereas AHIREM was not. The association between AHIREM/AHINREM and hypertension varies across the severity of OSA. Significant association to AHIREM was found in mild-moderate patients, whereas association to AHINREM was only shown in severe patients. This work was supported by the National Natural Science Foundation of China (81530002, 81629002, 81770087) .
What problem does this paper attempt to address?